--- Stefan Fent schrieb: > * Siegmund Langsch [040716

08:13]:

...

Hello,

Jul 15 20:59:31 LXK0D85B2 kernel: hda: drive_cmd: status=0x51 { DriveReady SeekComplete Error } Jul 15 20:59:31 LXK0D85B2 kernel: hda: drive_cmd: error=0x04 { DriveStatusError }

Check with hdparm wether DMA is is used. The output from 'hdparm /dev/hda' should contain a line: using_dma = 1 (on))

if not, switch it on with 'hdarm -d1 /dev/hda'.

I get the same error messages after during the boot; i receive them for all EIDE devices including CD Recorder and floppy drives. When i'm playing music CD this error is repeating hundreds times ... and yes, i'm using the DMA ;-) any ideas what could be the problem ? best regards, Franz PS my system: AMD Athlon64 3000+ MSI K8T Neo (VIA FIS2R chipset) GeForce 5700 FX GPU Seagate 120GB EIDE harddisk LG Electronics 52x52x24 CD Burner 3,5" floppy drive ___________________________________________________________ Gesendet von Yahoo! Mail - Jetzt mit 100MB Speicher kostenlos - Hier anmelden: http://mail.yahoo.de

Greetings, I'm poking my head into this thread because your slowness/ DMA timeout error symptoms look EXACTLY like a problem I had with an Athlon64 Laptop recently and Suse 9.1, with both the install and later update kernel... except mine didn't crash. There is a thread a bit back on "suse-amd64" whose title begins with "Suse 9.1 on my A64 laptop idles at huge load". I summarize below: =?iso-8859-1?q?Franz=20Mach?= wrote:

just forgot to say that i'm using SuSE 9.1 (AMD64) with the 2.6.5-7.75 kernel.

i'm also getting this error message during the startup :

...

Jul 15 20:59:31 LXK0D85B2 kernel: powernow-k8: BIOS error - no PSB

is this anything important ?

about slow and hot: my system isn't slow - after about 30minutes the fans start to work quite hard but it never goes too hot even after 10 hours ...

My symptoms were: Kernel from install: -- Powernow working. -- Processor jumped to max speed and stayed there (2.2GHz in my case). -- Some DMA errors showed up in the log. -- Load was 7-8, machine very very sluggish, but no obvious culprits when running top. If you just count the CPU time spent in individual processes in top/ps, the machine looked idle, but the CPU utilization summary reported 100%. Kernel update 2.6.5-7.95, changes: -- Powernow broken, claim of no PSB found. -- Runs at 800MHz. Summary: We found fixes for both problems. 1) Powernow broken in later kernel: This turned out to be a simple bug and we narrowed down to a small patch which I posted to the list near the end of the "Suse 9.1 laptop..." thread listed above. Essentially Powernow ACPI support was disabled by accident, and we turn it back on with the patch. 2) Huge load/slowness: This was a problem with the "powersave_proxy" script which is part of the "powersaved" daemon. There is a bad interaction with some ACPI BIOSes, including, unfortunately, the one on my laptop. A simple test to see if this is the case is to stop the powersave daemon with (as root): /etc/init.d/powersave stop In my case, the above got rid of the huge load. If this works for you, then if you look through the same email thread listed above, you'll find what I did to work around the problem but leave powersave support enabled. 3) My machine never crashed from the heat. I agree the huge load was probably just stressing your inadequate CPU cooling/heatsink solution. Sorry I can't be more helpful there... -- Erich Stefan Boleyn http://www.uruk.org/ "Reality is truly stranger than fiction; Probably why fiction is so popular"

Hello, Janis Klava wrote:

seti@home is probably a lot like D2OL in that it utilizes the processor close to 100% for maths intensive calculations. that is the reason for the slowness.

Thank you, but: That isn't it. The "slowness" is also present even seti is *not* running. In normal cases seti catches only *this* CPU-time which is not used by other processes and releases this time if another process wants it. By Siggi